Glovers reef survey in Belize, a multi-Aquarium project
La surveillance du récif de Glovers à Bélize,
un projet multi-Aquariums
D. A. THONEY
New York Aquarium, Wildlife Conservation Society, Brooklyn, NY 11224, USA
ABSTRACT
In 1994, the Wildlife Conservation Society established the Middle Cay Marine Research Station
in Belize at the southern end of Glover's Reef, one of the few atolls in the Caribbean. Opening the
station was an important step in creating the necessary infrastructure for conducting research needed
for managing the newly created Marine Reserve. In order to monitor Glover's Reef ecology and to
provide essential information for effective management decisions, a baseline faunal and flora survey
was conducted of the key reef inhabitants. The goals of the survey were to determine abundance and
distributions of algae, invertebrates especially corals, and fishes living in the patch reef area of
Glover's Reef (over 700 patch reefs; 312 sq. km), and to store and analyze the data in a geographic
information system (GIS) as the beginning of a long-term study of this important reef ecosystem.
Forty-four scientists from 22 public aquariums spent 722 man-days collecting information required
for the data base. One hundred eighty patch reefs were sampled, each consisting of five fish surveys
and four benthic transects. In addition, 112 bounce dives were conducted to collect sand and to
quickly access substrate between patch reefs. Future surveys will be conducted to study the
effectiveness of the reserve and to access changes that result from natural phenomena such as
hurricanes and extreme temperature fluctuations.
RÉSUMÉ
En 1994, la “Wildlife Conservation Society” a fondé la Station de Recherche Marine du “Middle
Cay” à Bélize, dans la partie sud du récif de Glovers, un des rares atolls des Caraïbes. L’ouverture
de la Station a représenté un progrès important dans la création des infrastructures nécessaires à la
conduite des recherches pour la gestion de la nouvelle réserve marine. De façon à surveiller le récif
de Glovers et à fournir des informations essentielles pour la prise de décisions en matière de gestion,
une étude de la faune et de la flore a été entreprise sur les peuplements clés du récif. L’objectif de
cette étude était de déterminer l’abondance et la distribution des algues, des invertébrés,
particulièrement les coraux, et des poissons qui vivent dans la zone des pâtés coralliens du cérif de
Glovers, composé de plus de 700 pâtés répartis sur 312 km², d’acquérir et de stocker les données
dans un système d’information géographique (SIG), pour débuter une étude sur le long terme de cet
important écosystème récifal. 44 scientifiques de 22 Aquariums publics ont passé près de 722 jours
de travail à collecter les informations nécessaires pour constituer la base de données. 180 pâtés
coralliens ont été échantillonés, chacun de ces échantillonnages consistant en 5 études des poissons
et 4 transects benthiques. De plus, 112 plongées ont permis de collecter du sable et de connaître
rapidement la nature du substrat entre les pâtés coralliens. Les futures missions de surveillance
devront étudier l’efficacité de la réserve et mesurer l’impact de perturbations naturelles telles les
ouragans ou les fluctuations extrêmes de température.
Bulletin de l’Institut océanographique, Monaco, n° spécial 20, fascicule 2 (2001)
INTRODUCTION
In 1994, the Wildlife Conservation Society established the Middle Cay Marine
Research Station in Belize at the southern end of Glover's Reef, one of the few
atolls in the Caribbean. Opening the station was an important step in creating the
necessary infrastructure for conducting research needed for managing the newly
created Marine Reserve. In order to monitor Glover's Reef ecology and to provide
essential information for effective management decisions, a baseline faunal and
flora survey was conducted of the key reef inhabitants. The goals of the survey were
to determine abundance and distributions of algae, invertebrates especially corals,
and fishes living in the patch reef area of Glover's Reef (over 700 patch reefs; 312
sq. km), and to store and analyze the data in a geographic information system (GIS)
as the beginning of a long-term study of this important reef ecosystem.
METHODS
Project Design and Implementation
The proposed faunal survey was carried out over a two-year period by
conducting six two-three week visits to Glovers Reef by seven-eight member
survey teams. Each team consisted of a supervisor and survey biologists. For every
patch reef sampled a benthic survey was conducted in N-S, W-E, S-N, and E-W
direction. In addition, five point fish counts were completed on the N, W, S, E and
center of the reef.
The atoll was divided into seven, approximately same-sized areas (each about 35
sq. km) (Figure 1), and each area was divided into 0.25 sq. nm quadrants. A
quadrant randomly selected from one of the areas was sampled each day. After
arriving at the GPS coordinates for a quadrant, three patch reefs were selected based
on three size categories, respectively. Once satellite or aerial photographs become
available, locations will be matched exact reefs depicted in the photograph. Once
selected, each patch reef was be surveyed for algae, corals, and fish.
Figure 1: Glovers reef atoll, Belize
Bulletin de l’Institut océanographique, Monaco, n° spécial 20, fascicule 2 (2001)
The «line intercept transect method» was used to survey sessile invertebrate
organisms (algae and corals) on patch reefs. This technique was first developed for
terrestrial plant ecology and was adapted to coral reef studies by Loya (1978) and
Marsh et al. (1984). A ten meter long weighted taped marked in 10 cm intervals
was loosely laid over the substrate. Algae and invertebrates dominating each 10 cm
interval were coded and recorded on standardized forms attached to underwater
slates as the diver slowly swam the transect. To calculate percentage of cover, the
total length occupied by a species along the transect was divided by the total length
of the transect. Those percentages were then averaged with the other transects on
that patch reef to determine area coverage by each species for that patch (English et
al., 1997). Principle component analysis was used to determine species patterns
between areas.
A visual census survey of reef fish communities was conducted in the patch reef
environment of Glovers Reef Atoll. Sites for fish analyses were at the same patch
reefs that were selected for the benthic transects (see above). Each census was
conducted at the three meter contour of the patch reef using SCUBA and/or
snorkeling following the method of Bohnsack and Bannerot (1986). This method
has proven to be suitable for describing reef fish community structure, and for
making comparisons of fish abundance and community structure among sites and
habitats (Bohnsack and Talbot, 1980; Bohnsack, 1982; Bohnsack and Bannerot,
1986; Clavijo et al., 1989). This method consisted of a diver listing the fish species
seen in an imaginary five meter radius cylinder of water in a five-minute period at
one of the selected sites. Following the initial five-minute census, the diver spent 20
minutes recording counts or estimates of abundance, and minimum and maximum
length for only those species observed in the first five-minute census. Any new
species that entered the cylinder area after the initial five minute survey were
excluded. Data was recorded on standardized forms attached to underwater slates
and tallied at the end of each day. Principle component analysis was used to
determine species patterns between areas.
Using biologists from aquariums that work with many of these species day-inand-day-out saved numerous days of training and improved the integrate of the data
significantly. Forty-four biologists from 22 public aquariums spent 722 man-days
collecting information required for the data base. One hundred eighty patch reefs
were sampled, each consisting of five fish surveys and four benthic transects. In
addition, 112 bounce dives were conducted to collect sand and to quickly access
substrate between patch reefs. Several water samples also were collected to study
nutrient distributions across the reef. The bounce dives and nutrient data will not be
discussed in this paper.
RESULTS
The project resulted in the establishment of a
GIS
database for Glovers Reef,
Bulletin de l’Institut océanographique, Monaco, n° spécial 20, fascicule 2 (2001)
with distributions and relative abundance of key algae, corals, and fishes recorded
for the patch reef area. Sixty-four species of fish, 28 species of hard coral, 10
genera of soft coral, 10 common taxa of other invertebrates, and 20 algae were
identified.
Benthic Data
The Benthic communities in Area 6 are different from the other two correlated
groups (2,3 and 4,5), but more similar to Area 7. Factor 1 accounted for 23.6% of
the variance and was mostly associated with species of coral. Factor 2 accounted
for 17.9% of the variance and was mostly associated with macro-algae &
invertebrates. Factor 3 accounted for 8.3% of the variance and mostly associated
with Lobophora sp. and other macro algae.
Cluster Analysis comparing areas (Figure 2) showed that the benthic
communities in Areas 1 & 2, 3 & 4, and 5 & 6 form three distinct groups, and that
the three groups are related to each other at similar levels. Area 7 is grouped quite
separately from the rest of the areas.
Figure 2: Distribution of communities of sessile organisms in seven areas
Cluster Analysis comparing benthic species and substrate type relationships
showed that most of the invertebrates, including corals and the smaller red algae
group together separate from the larger brown algae, Dictyota, Sargassum,
Turbinaria, Padina and the finger coral, Porites porites. The common brown
macroalgae Lobophora was grouped separately and the sea grass Thalasia was the
furthest separated.
When major taxa and substrate type are compared between areas, it is apparent
that macro-algae dominate other benthic organisms across the entire interior of the
atoll with the highest concentrations occurring in Areas 2, 3, 4, 5, & 6. Hard coral
coverage is relatively low throughout the atoll, but highest in Areas 4, 5, 6, & 7, and
Bulletin de l’Institut océanographique, Monaco, n° spécial 20, fascicule 2 (2001)
percent coverage of soft corals appears low in most areas except Area 7 at the
southern end of the atoll. Sponges tend to be more common in Areas 3 & 7 where
turbidity appeared to be higher. Area 7 had the highest percentage of un-colonized
(by macro-organisms) coral rock. Sand and sea grasses did not appear to be
correlated and varied from area to area.
Fish Data
Principle Component Analysis indicates that the fish communities in Areas 1 &
2 and 3 & 4 are quite similar, respectively, and that Areas 5 & 6 are more similar
than they are to Area 7. Factor 1 accounted for 23.6% of the variance and was
mostly associated with lutjanids (snappers) and pomadasyids (grunts). Factor 2
accounted for 13.2% of the variance and was mostly associated with pomadasyids,
scarids (parrotfish), lutjanids, and Pomacanthus (french angel). Factor 3 accounted
for 8.8% of the variance and was mostly associated with serranids (groupers) and
acanthurids (tangs).
Cluster Analysis comparing areas (Figure 3) shows that the fish communities in
Areas 3, 4, 5, & 6 are similarly related to each other. Areas 1 & 2 are more similar
to that group than they are to group 7, which groups out quite differently.
Figure 3: Distribution of communities of fishes in seven areas
Cluster Analysis comparing fish species relationships shows that most of the
coral reef fishes including the acanthurids, chaetodontids (butterflyfishes), labrids
(wrasses), pomacanthids (angelfishes), pomacentrids (damselfishes), pomadasyids,
serranids, sparids (porgies) and others group out together. However, the following
groups: 1) yellow wrasse, white grunt, and redfin parrotfish, 2) bar jack, and 3) the
tomtate and lane snapper are more distally related from previous groups in
ascending order. The major families of fishes occur in all seven areas, however,
patterns are apparent. Firstly, The labrids and scarids outnumber all of the other
Bulletin de l’Institut océanographique, Monaco, n° spécial 20, fascicule 2 (2001)
families. Secondly, the populations of acanthurids, pomacanthids, and serranids are
larger in Area 7, than over the rest of the atoll. And thirdly, pomadasyids are more
common in Areas 1, 2, 3, 4, & 5 than the southern portion of the atoll.
Fish and Benthic Relationships
Several species representing the Acanthuridae, Serranidae, Labridae, and others
were found to be significantly associated with various species of coral. In addition,
some of the same species and others including several acanthurids, labrids,
pomacentrids, and scarids, were negatively associated with macro-algae.
DISCUSSION
In 1974, Wallace described the patch reefs of Glovers Atoll as being dominated
by several species of coral including large outcroppings of Acropora palmata.
Since that time, A. palmata and A. cervicornis colonies have been decimated
through out the Caribbean by White Band Disease, but that does not explain the
loss of other species of coral that once dominated this atoll. Many of them are now
overgrown by algae. Overgrowth of corals by algae appears to be a relatively recent
phenomena that has occurred in the last 10-20 years. Large areas are covered with
dead skeletons of Porites porites, Acropora cervicornis, Diploria spp., and
Montastrea annularis that are still identifiable even though they have been battered
by waves and currents. They now provide substrate for algae, sponges, and
tunicates. On a more positive note, however, acroporid corals appear to be making a
slow come back.
Patch reefs in the center of the atoll have significantly higher abundance of
macro-algae and fewer healthy corals than along the edge of the atoll or where there
are cuts in the atoll, which may be related to current and wave energy. Although
major differences in algae growth and diversity patterns between patch reefs
appears to be related to current and wave energy, nutrients also may be a factor.
Fewer fish are associated with these algae dominated reefs. Angelfish and
groupers are more common in the northern and southern portions of the reef where
the algae is less common. Herbivores such as tangs, parrotfish, and chub are more
highly correlated with coral than they are with macro-algae. The large brown algae
that dominate these reefs are not grazed upon by these fishes.
In contrast, grunts are not only more common in the middle reaches of the reef,
but more highly correlated with macro-algae. Grunts and snappers often feed in
grass beds on invertebrates and small fishes. The macro-algae may provide them
with similar habitat requirements for feeding and protection from predators.
It is not clear what has allowed the macro-algae to dominate many of the patch
Bulletin de l’Institut océanographique, Monaco, n° spécial 20, fascicule 2 (2001)
reefs within Glovers Atoll. It is likely that the lack of herbivores may be having a
significant affect on the patch reefs. Most of the Caribbean suffered from a
catastrophic die off of Diadema urchins, which are significant grazers of algae. In
addition, personal communications with local fishermen and divers suggest that
there were large schools of rainbow and midnight blue parrotfish that have
disappeared over the last decade. Not one rainbow parrotfish was identified during
the survey work. Overfishing is most likely the cause of their disappearance. With
fewer grazers present macro-algae are able to establish themselves, and once
established they are not eaten by any of the common fishes or invertebrates.
Nutrients may play a factor, but it is not clear how. The Atoll is too far offshore
to be affected by anthropogenic pollution, and upwelling has been a common
occurrence all along the atoll for millennia. Therefore, nutrient loading should not
have increased in recent times. However, nutrient concentrations are probably
naturally high on the reef and may contribute to the establishment of macro-algae.
Further sampling will be necessary in order to discern any patterns relating to
nutrients.
Future surveys will be conducted to study the effectiveness of the reserve and to
access changes that result from natural phenomena such as hurricanes and extreme
temperature fluctuations.
ACKNOWLEDGMENT
I wish to thank all my colleagues that participated in the field research from the
following institutions: Aquarium of the Americas, Columbus Zoo, New Jersey
State Aquarium, Florida State Aquarium, Fort Worth Zoo, John G. Shedd
Aquarium, Humboldt State University, Monterey Bay Aquarium, Mystic Aquarium,
New England Aquarium, Pittsburgh Zoo, Sea World of Ohio, Sea World Of Texas,
Toledo Zoo, and Cleveland Metroparks Zoo. A project of this size could not have
occurred successfully without their help. I also wish to thank my research my
technician, Samual Jones, for his untiring effort in many aspects of this project.
This work was supported by Conservation Fund, Wildlife Conservation Society and
the Conservation Endowment Fund, American Zoo and Aquarium Association.
REFERENCES
BOHNSACK, J. A., 1982.- Effects of piscivorous predator removal on coral reef fish
community structure; Fish food habits studies-. Gutshop. N° 81. Proceedings of
the Third Pacific Workshop. Washington Sea Grant Publication WSG-WO.
Eds.G. M. Cailliet, and C. A. Simenstad. 82-2. 258-267 pp.
Bulletin de l’Institut océanographique, Monaco, n° spécial 20, fascicule 2 (2001)
BOHNSACK, J. A., BANNEROT S. P., 1986.- A stationary visual census technique for
quantitatively assessing community structure of coral reef fish-. NOAA
Technical Report. NMFS 41. 15 p.
BOHNSACK, J. A., TALBOT F. H., 1980.- Species packing by reef fishes on
Australian and Caribbean reefs; an experimental approach-. Bulletin of Marine
Science. 30 (3): 710-723.
CLAVIJO, I. E., LINDQUIST D. G., BOLDEN S. K., and BURK S. W., 1989.- Diver
inventory of a midshelf reef fish community in Onslow Bay, North Carolina:
preliminary results for 1988 and 1989-. Diving for science. Eds M. A. Lang, and
W. C. Jaap. American Academy of Underwater Science. 59-65.
ENGLISH, S., C. WILKINSON, and V. BAKER., 1997.- Survey Manual for Tropical
Marine Resources. 2nd edition. Australian Institute of Marine Science.
Townsville, Australia.
LOYA, Y., 1978.- Plotless and transect methods-. Coral Reefs: Research Methods.
Eds Stoddart, D. R. and R. F.Johannes. UNESCO, Paris. 197-217.
MARSH, L. M.,. BRADBURY R. H, and REICHELT R. E., 1984.- Determination of the
physical parameters of coral distributions using line transect data.-. Coral Reefs.
2: 175-180.
WALLACE, R. J., 1974.- A reconnaissance of the sedimentology and ecology of
Glover’s Reef Atoll, Belize-. Ph.D. Dissertation, Princeton
University.Princeton, New Jersey.
Bulletin de l’Institut océanographique, Monaco, n° spécial 20, fascicule 2 (2001)